No place to hide: Refugees, displaced persons, and the recruitment of child soldiers

International Security

Volumn 31, Issue 1, 2006, Pages 127-164

  Achvarina, Vera ^a   Reich, Simon F ^a,b  

^a UNIVERSITY OF PITTSBURGH   (United States)

^b GSPIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 33747168772     PISSN: 01622889     EISSN: 15314804     Source Type: Journal    
DOI: 10.1162/isec.2006.31.1.127     Document Type: Article

References (232)

1. See, for example, "U.S. Commander: Taliban Recruiting Children," July 23, 2005, http://www.cnn.com/2005/WORLD/asiapcf/07/23/ afghan.taliban.ap/.
2. UN Office for Coordination of Humanitarian Affairs, "Afghanistan: Eight Thousand Children Under Arms Look for a Future," http://www.irinnews. org/webspecials/childsoldiers/Afghanistan031203.asp#lop.
3. See "Child Suicide Bombers in the Israeli-Palestinian Conflict," http://en.wikipedia.org/wiki/Child_suicide_bomber;
4. Human Rights Watch, "Erased in a Moment: Suicide Bombing Attacks against Israeli Civilians," http://www.hrw.org/reports/2002/isrl-pa/ ISRAELPA1002.pdf;
5. and Children and Armed Conflict Unit, Defense for Children International, Palestine Section, "Use of Children in the Occupied Palestinian Territories: Perspective on Child Soldiers," July 2004, http://www.essex.ac.uk/armedcon/story_id/00205.pdf.
6. Center for Defense Information, "Children on the Front Line: Child Soldiers in Afghanistan," October 15, 2001, http://www.cdi.org/ friendlyversion/printversion.cfm?documentID?2038.
7. Ibid. "Children on the Front Line: Child Soldiers in Afghanistan," 2001, http://www.cdi.org/friendlyversion/printversion.cfm? documentID?2038.
8. In this article we use the United Nations Children's Fund's definition of a child soldier: "any child - boy or girl - under 18 years of age, who is part of any kind of regular or irregular armed force or armed group in any capacity, including, but not limited to, cooks, porters, messengers, and anyone accompanying such groups other than family members." See UNICEF, "Fact Sheet: Child Soldiers," http://www.unicef.org/protection/childsoldiers. pdf, p. 4.
9. For the 1988 estimate of 200,000 child soldiers, see "Children in War," in UNICEF, The State of the World's Children: 1996, http://www.unicef.org/sowc96/2csoldrs.htm.
10. See also "UN Cites Child Recruiters but Omits Leading Offenders," Human Rights News, December 16, 2002, http://www.hrw.org/press/ 2002/12/childsoldiers1216.htm.
11. See Coalition to Stop the Use of Child Soldiers (CSUCS), Child Soldiers Global Report, 2004, February 2004, http://www.child-soldiers.org/document_get. php?id=966, pp. 13-17.
12. This claim is based on the figures discussed in detail elsewhere in this article. For examples of such wars, see Jo Becker, "Children as Weapons of War," in Human Rights Watch, ed., World Report, 2004: Human Rights and Armed Conflict (New York: Human Rights Watch, 2004), pp. 219-244.
13. See, for example, Coral Bell, "Normative Shift," National Interest, No. 70 (Winter 2002/03), pp. 44-54;
14. Martha Finnemore, "Constructing Norms of Human Intervention," in Peter J. Katzenstein, ed., The Culture of National Security: Norms and Identity in World Politics (New York: Columbia University Press, 1996), p. 158;
15. and Ronald L. Jepperson, Alexander Wendt, and Peter J. Katzenstein, "Norms, Identity, and Culture in National Security," in Katzenstein, The Culture of National Security: Norms and Identity in World Politics, p. 45.
16. Currently, the most important formal international conventions and protocols related to recruitment of children in armed conflicts are: "Geneva Convention," Additional Protocol I (relating to international armed conflicts), art. 77(2); Additional Protocol II (relating to non-international armed conflicts), art. 4(3)c; "Convention on the Rights of the Child," art. 38; "Rome Statute for an International Criminal Court," art. 8 (on war crimes), sec. 2(b) (xxvi); Optional Protocol to the Convention on the Rights of the Child on the Involvement of Children in Armed Conflict, art. 4(1) and (2); and the "African Charter on the Rights and Welfare of the Child." For this list, see Coalition to Stop the Use of Child Soldiers, "Child Soldiers 1379 Report," November 2002, http://www.child-soldiers.org/document_get.php?id=740, pp. 7, 8.
17. P.W. Singer, Children at War (New York: Pantheon, 2005), pp. 15, 38.
18. The Ugandan conflict is described in the literature as "a war fought by children on children," where "minors make up almost 90% of the Lord's Resistance Army's soldiers." According to the United Nations, "Since the rebellion began in the 1980s, some 30,000 children have been abducted to work as child soldiers and porters, or to serve as 'wives' of rebels and bear their children." "Uganda: Child Soldiers at Centre of Mounting Humanitarian Crisis," Ten Stories the World Should Hear More About, 2004, http://www.un.org/events/tenstories/story.asp?storyID=100. According to UNICEF
19. "[an estimated] 8,400 children were abducted between June 2002 and May 2003. In July 2003 more than 20,000 child 'night commuters' were estimated to seek safety each night in Gulu, Pader, and Kitgum towns, to reduce the risk of abduction." CSUCS, Child Soldiers Global Report, 2004, p. 106.
20. See UNICEF, Adult Wars, Child Soldiers (Bangkok: UNICEF, October 2002), http://www.unicof.org/emorg/AdultWarsChildSoldiors.pdf, p. 8.
21. These distributions were calculated on the basis of data found in United Nations High Commissioner for Refugees, "Liberia: What Hope for Peace?" October 1, 1994, http://www.unhcr.org/cgi-bin/texis/vtx/print?tbl= RSDCOI&id=3ae6a6bc0;
22. Fiona Callistor, "Liberia's Child Soldiers," Tablet (London), October 4, 2003, http://www.thetablet.co.uk/cgi-bin/archive_db.cgi?tablet-00789;
23. and Amnesty International, "Liberia: Demand Justice for Child Soldiers," May 17, 2004, http://web.amnesty.org/pages/lbr-170504-action- eng.
24. "Sudan Protests UNICEF Child Soldier Airlift," CNN.com, March 7, 2001;
25. Rädda Barnen, child war database, http://www.globalmarch.org/ resourcecenter/world/sudan.pdf;
26. and CSUCS, Child Soldiers Global Report, 2004.
27. These figures were calculated on the basis of data drawn from CSUCS, "Child Soldiers 1379 Report";
28. Human Rights Watch, "Forgotten Fighters: Child Soldiers in Angola," Vol. 15, No. 10(A) (April 2003), http://www.hrw.org/reports/2003/ angola0403/;
29. and CSUCS, Child Soldiers Global Report, 2004.
30. This number was derived by analyzing the following sources: CSUCS, "Child Soldiers 1379 Report";
31. Human Rights Watch, "Forgotten Fighters";
32. and CSUCS, Child Soldiers Global Report, 2004.
33. Eleven countries in East Asia and the Pacific had child soldier participants. Laura Barnitz, Child Soldiers: Youth Who Participate in Armed Conflict, 2d ed. (Washington, D.C.: Youth Advocate Program International, 1999), pp. 2-3.
34. Coalition to Stop the Use of Child Soldiers, "The Use of Children as Soldiers in Africa: A Country Analysis of Child Recruitment and Participation in Armed Conflict," August 2000, http://www.reliefweb.int/library/ documents/chilsold.htm;
35. Afua Twum-Danso, Africa's Young Soldiers: The Co-option of Childhood, Monograph, No. 82, Institute for Security Studies, April 2003, http://www.iss.co.za/Pubs/Monographs/No82/Content.html, p. 8.
36. See UNICEF, Adult Wars, Child Soldiers.
37. Evidence to support this claim is extensive. Reportedly, child soldier numbers in Uganda and the Democratic Republic of the Congo, for example, increased dramatically during 2002 and 2003. See Becker, "Children as Weapons of War," p. 219. Observers also claim that there was a massive increase in recruitment in the Ivory Coast in 2003.
38. See Coalition to Stop the Use of Child Soldiers, "Child Soldier Use, 2004: A Briefing for the 4th UN Security Council, Open Debate on Children and Armed Conflict," 2003, http://www.hrw.org/reports/2004/childsoldiers0104/ index.htm.
39. See Jo Becker and Tony Tate, Stolen Children: Abduction and Recruitment in Northern Uganda (New York: Human Rights Watch, 2003), http://www.hrw.org/ reports/2003/uganda0303/uganda0403.pdf, p. 7.
40. The authors report that in Uganda the age of children being abducted had fallen from the 13-15 range to as low as 9 or 10. When child soldiers were demobilized at the conclusion of hostilities in Mozambique, 4,678 of all officially demobilized children (or 18 percent) were younger than 13 when recruited; 6,829 (27 percent) were 14-15 years old; and 13,982 (55 percent), were 16-17 years old. "Africa: After the Wars," http://www.cbc.ca/ afterthewars/childsoldier.html.
41. Singer, Children at War, pp. 38, 55.
42. Office of the Special Representative of the Secretary-General for Children and Armed Conflict, "Child Soldiers," http://www.un.org/ special-rep/children-armed.-conflict/English/ChildSoldiers.html.
43. Guy Goodwin-Gill and Ilene Cohn, Child Soldiers: The Role of Children in Armed Conflict (New York: Clarendon, 1994), pp. 23, 31, 38.
44. See also, for example, Twum-Danso, Africa's Young Soldiers.
45. See, for example, Barnitz, Child Soldiers, p. 23;
46. Twum-Danso, Africa's Young Soldiers, pp. 8, 10;
47. Graça Machel, "Impact of Armed Conflict on Children," August 26, 1996, http://www.unicef.org/graca/, p. 11;
48. Coalition to Stop the Use of Child Soldiers, "Child Soldier Use, 2003: A Briefing for the 4th UN Security Council, Open Debate on Children and Armed Conflict," January 2004, http://www.hrw.org/reports/2004/ childsoldiers0104/childsoldiers.pdf, p. 2;
49. and Coalition to Stop the Use of Child Soldiers, Child Soldiers Global Report, 2001, http://www.reliefweb.int/rw/lib.nsf/db900SID/LGEL-5CSHGX? OpenDocument.
50. For this argument, see Shannon McManimon, "Use of Children as Soldiers," Foreign Policy in Focus, Vol. 4, No. 27 (November 1999), http://www.fpif.org/briefs/vol4/v4n27child_body.html.
51. In the Liberian civil war, for instance, factions avoided fighting as much as possible, so that simulated attacks could facilitate looting. See Stephen Ellis, The Mask of Anarchy: The Destruction of Liberia and the Religious Dimension of an African Civil War (London: Hurst, 1999), p. 145.
52. Singer, Children at War, pp. 9-15.
53. Quaker United Nations Office, "Child Soldiers: Why Adolescents Volunteer," oral statement by Friends World Committee for Consultation (the Quakers) to the sixtieth session of the UN Commission on Human Rights, April 7, 2004, http://www.geneva.qu.no.info/pdf/CHR60ChildSoldiersoral.pdf (emphasis in original).
54. Both Twum-Danso and Barnitz derive the significance of orphans as a determinant of child soldier rates from Rachel Brett and Margaret McCallin, Children: The Invisible Soldiers (Växjö, Sweden: Rädda Barnen, 1996).
55. Human Rights Watch, "Stop the Use of Child Soldiers! U.S. Congress Condemns the Use of Children," http://hrw.org/campaigns/ap/congress.htm.
56. See, for example, UNICEF, Adult Wars, Child Soldiers.
57. See Goodwill-Gill and Cohn, Child Soldiers, pp. 23, 31, 38;
58. Twum-Danso, Africa's Young Soldiers;
59. and Barnitz, Child Soldiers, pp. 2-3.
60. Singer, Children at War, p. 62.
61. For examples of narratives describing the recruitment of orphans that illustrate this point, see Human Rights Watch, "Easy Prey: Child Soldiers in Liberia," 1994, http://www.hrw.org/reports/pdfs/c/crd/liberia949.pdf, pp. 15-19;
62. and UNICEF, Adult Wars, Child Soldiers.
63. UNICEF, Adult Wars, Child Soldiers, p. 30.
64. Carnegie Commission on Preventing Deadly Conflict, Preventing Deadly Conflict: Small Arms Survey, 2002 (New York: Oxford University Press, 2002), pp. 203-231;
65. Keith Krause, "Multilateral Diplomacy, Norm Building, and UN Conferences: The Case of Small Arms and Light Weapons," Global Governance, Vol. 8, No. 2 (April-June 2002), pp. 247-263;
66. and Thomas Jackson, Nicholas Marsh, Taylor Owen, and Anne Thurin, Who Takes the Bullet? The Impact of Small Arms Violence, Understanding the Issues series, No. 3 (Oslo: Norwegian Church Aid, 2005). For more background on the small arms issue, including the United Nations 2001 Program of Action, see http://www.iansa.org, http://www.smallarmssurvey.org, and http://www.nisat.org.
67. Singer, Children al War, p. 38.
68. McManimon, "Use of Children as Soldiers."
69. Singer, Children at War, p. 47.
70. Ibid., Singer, Children at War, p. 16.
71. For a discussion of how children played this role in the first Liberian conflict, see Human Rights Watch, "Easy Prey," pp. 32-34.
72. For example, in Child Soldiers, Goodwin-Gill and Cohn mention a variety of explanations that are hard to operationalize. On page 31, for instance, they define the militarization of daily life as the "presence of heavily armed policemen or soldiers patrolling the streets, military personnel occupying high government posts, military censorship of social life, armed guards in schools and public buildings, armed checkpoints along the roads, and curfews." Comparably, the link between children's experiences of physical violence and their desire to take up arms is characterized as a "desire for revenge, conviction to continue the struggles of lost loved ones, the need to substitute an annihilated family or social structure, and the desire to take control over events that shape one's circumstances" (p. 32). In reference to "structural violence," the authors focus on broad social and economic injustices. They link these injustices to a child's motivation to obtain food (pp. 32-33). All of these terms are difficult to operationalize. For other comparable examples, see Barnitz, Child Soldiers;
73. Carolyn Nordstrom, Girls and War Zones: Troubling Questions (Uppsala, Sweden: Life and Peace Institute, May 1997);
74. and Twum-Danso, Africa's Young Soldiers.
75. For examples based on anecdotal evidence, see UNICEF, "Impact of Armed Conflict on Children";
76. UNICEF, Adult Wars, Child Soldiers;
77. Twum-Danso, Africa's Young Soldiers;
78. Barnitz, Child Soldiers;
79. and Goodwin-Gill and Cohn, Child Soldiers. The UNICEF report is based entirely on the interviews of sixty-nine former and current child soldiers. Both Twum-Danso and Barnitz identify several possible causal factors, but neither qualitatively nor quantitatively tests them. Based on the statements of children, the UNICEF study Adult Wars, Child Soldiers, for example, suggests that fear plays a potential role in the recruitment process. It does not, however, develop or test the argument.
80. See "Child Soldiers: An Overview," in CSUCS, Child Soldiers Global Report, 2001.
81. For examples, see Becker and Tate, Stolen Children;
82. Human Rights Witch, "Forgotten Fighters";
83. Human Rights Watch, "How to Fight, How to Kill: Child Soldiers in Liberia," Vol. 16, No. 2(A) (February 2004), http://hrw.org/reports/2004/ liberia0204/liberia0204.pdf;
84. Human Rights Watch, "Easy Prey";
85. and Alfred B. Zack-Williams, "Child Soldiers in the Civil War in Sierra Leone," Review of African Political Economy, Vol. 28, No. 87 (March 2001), pp. 73-82.
86. For a general discussion of the comparative utility of general and contingent theories, see, for example, Alexander George and Richard Smoke, Deterrence in American Foreign Policy: Theory and Practice (New York: Columbia University Press, 1974), appendix.
87. Later in this article, we offer systematic tests of the hypotheses about poverty and orphans. But we cannot do the same for the light arms argument given the lack of suitable data. Nonetheless, recognizing the importance of the argument, it is one we intend to return to in future research if possible.
88. We observe that the relevant literature suggests that camps for IDPs really started appearing on a widespread basis in these conflicts in the late 1990s, whereas refugee camps were widespread many years earlier.
89. See United Nations High Commissioner for Refugees, "Refugee Children in Africa: Trends and Patterns in the Refugee Population in Africa below the Age of 18 Years, 2000," June 2001, http://www.unhcr.org/cgi-bin/texis/vtx/ statistics/opendoc.pdf?tbl=STATISTICS&id=3b9378e42d.
90. Barnitz, Child Soldiers, p. 4.
91. Goodwin-Gill and Cohn, Child Soldiers, p. 32.
92. See, for example, Stephen John Stedman and Fred Tanner, "Refugees as Resources in War," in Stedman and Tanner, eds., Refugee Manipulation: War, Politics, and the Abuse of Human Suffering (Washington, D.C.: Brookings, 2003), pp. 1-15.
93. For an example of a study on food aid delivery that does emphasize the importance of protection, see Bill Frelick, "Assistance without Protection: Feed the Hungry, Clothe the Naked, and Watch Them Die," in Worldwide Refugee Information, 1997, http://www.iconet.org/IFA/ifa_assistance.htm;
94. and Fiona Terry, Condemned to Repeat? The Paradox of Humanitarian Action (Ithaca, N.Y.: Cornell University Press, 2002).
95. One of the few examples is Barnitz, Child Soldiers, p. 4.
96. CSUCS, Child Soldiers Global Report, 2004, pp. 31, 33.
97. Terry, Condemned to Repeat? p. 28.
98. Children constitute 57 percent of all inhabitants of UNHCR-mandated refugee camps. United Nations High Commissioner for Refugees, "Refugee Children in Africa."
99. Confidential interview, UNICEF official, New York, April 6, 2005.
100. Confidential interview, Save the Children official, Pittsburgh, Pennsylvania, March 25, 2005.
101. Stedman and Tanner, "Refugees as Resources in War," p. 4.
102. Ibid., Stedman and Tanner, "Refugees as Resources in War," p. 3.
103. Among our cases, we included some conflicts that commenced, before 1975, provided that they concluded during or after 1975 and that suitable data were available. Conflicts that ended prior to 1975 were omitted due to a lack of suitable data.
104. Model United Nations of the University of Chicago, "Reintegration of Child Soldiers," 2003, http://munuc.org/munucxvi/2003pdf/UNICEF_A.pdf, p. 5.
105. By "conflict episode," wo mean a unit or observation defined as "a conflict, a sub-conflict, or a subset of either over a period of time." See Håvard Strand, Lars W. Wilhelmsen, and Nils Petter Gleditsch, in collaboration with Peter Wallensteen, Margarita Sollenberg, Mikael Eriksson, Halvard Buhaug, and Jan Ketilrod, "Armed Conflict Datasot Codebook," ver. 1.1 (September 2002), http://www.prio.no/cwp/armedconflict/ old/v1_10/Codebook_v1_1.pdf. In adopting their approach, each conflict in their dataset "is likely to include several observations." "Armed conflict" is defined as "a contested incompatibility that concerns government and/or territory where the use of armed force between two parties, of which at least one is the government of a state, results in at least twenty five battle-related deaths" (p. 2). A "sub-conflict" must satisfy one or more of the following criteria: (1) ten continuous years with fewer than twenty-five battle-related deaths per year; (2) a change in the conflict type from internal conflict to internationalized internal conflict or vice versa; and (3) a complete change in the belligerents in a conflict (p. 3).
106. An "internal conflict" in Strand et al.'s dataset is defined as "being within a country between a government and one or more opposition groups, with no interference from other countries." Håvard Strand, Lars W. Wilhelmsen, and Nils Petter Gleditsch, in collaboration with Peter Wallensteen, Margarita Sollenberg, Mikael Eriksson, Halvard Buhaug, and Jan Ketilrod, "Armed Conflict Datasot Codebook," Ver. 1.1 Ibid., p. 8.
107. Strand et al. define an "internationalized internal conflict" as one in which both sides in the conflict receive support from other governments. Ibid. Håvard Strand, Lars W. Wilhelmsen, and Nils Petter Gleditsch, in collaboration with Peter Wallensteen, Margarita Sollenberg, Mikael Eriksson, Halvard Buhaug, and Jan Ketilrod, "Armed Conflict Datasot Codebook," Ver. 1.1
108. Between 1975 and 2002, however, there were only 8 international interstate conflicts in Africa.
109. This way of clustering results (by merging episodes with different actors) was necessitated by the nature of the data available on child soldiers. The data do not differentiate between various factions fighting at the same time or between two separate episodes in adjacent years.
110. From this point in the article, we refer to intrastate conflict as a cluster of episodes of what Strand et al., "Armed Conflict Datasot Cookbook," call internal and internationalized internal conflicts.
111. The three conflicts were Angola (1975-2002), South Africa (1966-93), and Sudan (1983-2002).
112. This conflict took place in Burundi (1993-94). For the reference on ethnic violence during these years, see "Burundi Civil War," http://www.globalsecurity.org/military/world/war/burundi.htm.
113. We also had to alter the dates of one conflict in our sample. Although Strand et al. list the Liberian conflict as resuming in 2000, the literature on civil wars in Liberia dates the second conflict from as early as 1997. See, for example, "Liberia-Second Civil War-1997-2003," http://www. globalsecurity.org/military/world/war/liberia-1997.htm. The same source, however, reports on the first instances of violence emerging in 1999 in the form of insurgencies. Strand et al.'s dataset includes conflicts up to 2002, but the Liberian conflict continued into 2003. We therefore list the duration of this conflict as 1999 to 2003.
114. The number of children being recruited was estimated from a series of reports with figures compiled by different operating field organizations including CSUCS, Human Rights Watch, Save the Children, and UNICEF.
115. We code "minor use" as involving no evidence of systematic attempts to recruit children, no large-scale usage of them in conflict or as auxiliaries, and no competition to engage them by the government and rebel factions. The occasional report of a child in or near combat does not constitute the basis for inclusion as a case.
116. These seven conflicts were concentrated in five countries, with Niger being the location for three of them, and Mali and Senegal two each.
117. Although less reliable, the duration of a conflict may also serve as a proxy for the age distribution of a population as a possible explanation of child soldier rates. In principle, the percentage of child soldiers may rise as a war progresses because as adults die in conflict, children constitute a higher percentage of the population. We found no evidence to support this claim.
118. If the governmental forces were excluded from the calculation of child soldier ratios for the Ugandan conflict (1994-2002), the ratio would be 71 percent instead of the 22 percent that we have estimated. Lisa Sekaggya, program coordinator for social protection at Save the Children in Uganda, estimates that the child soldier figure for oppositional forces is 90 percent. See Sekaggya, "Ugandan Children Born in Captivity and Their Human Rights," paper presented at "Conference on War Babies," Pittsburgh, Pennsylvania, November 13, 2004.
119. As a result of the limited number of cases tested, we make no definitive claims on the basis of the analysis presented here. The findings are meant to be suggestive and, in the spirit of the comments made in some authoritative studies, we believe that quantitative analysis based on a limited N, if genuinely exhaustive and recognized as a plausibility probe, is better than none at all. For a defense of the use of regression analysis, even when one has a relatively limited N, see Gary King, Robert O. Keohane, and Sidney Verba, Designing Social Inquiry: Scientific Inference in Qualitative Research (Princeton, N.J.: Princeton University Press, 1994);
120. and Barbara Geddes, Paradigms and Sand Castles: Theory Building find Research Design in Comparative Politics (Ann Arbor: University of Michigan Press, 2003).
121. World Bank Group, Poverty Manual, August 8, 2005, http://siteresources. worldbank.org/
122. PGLP/Resources/povertymanual_ch3.pdf, chap. 3, p. 42. Poverty can be measured in several ways. Alternative methods are listed at the World Bank's PovertyNet, http://www.worldbank.org/poverty/mission/up1.htm;
123. and the Human Development indicators, http://hdr.undp.org/reports/global/ 2003/pdf/hdr03_HDI.pdf. We have chosen percentage of population below the poverty line because this measure allows us to focus on income and consumption levels as indicators of the degree of economic need.
124. Poverty data were unavailable for Liberia (1989-95), Lesotho (1998), and Mali (1994). Mali was excluded from our sample; in Lesotho, however, the poverty measure available was for the year after the conflict concluded and is included in the sample. Furthermore, we address the absence of data on poverty in the middle of the nine-year conflict in Uganda by calculating the average figure between the poverty rate at the outset and at the end of the conflict. We believe that this average represents a balanced characterization of the degree of poverty during the conflict. In the case of Angola (1975-95), data regarding the percentage of the population living below the poverty line were available only for rural areas. According to the Institute for Security Studies' population and social indicators database, http://www.iss.org.za/AF/profiles/ angola/Table_Population.html# Anchor-43556, the urban population of Angola was 17.8 percent in 1975 and 32.3 percent in 1997. The average of these numbers (25 percent) represents the urban population during the conflict of 1975-95. Given that the rural population of Angola made up 75 percent of the country's population during the conflict, we incorporated only the rural value of poverty as a proxy for poverty measurement in this case. Although this risks skewing the data, we justify including Angola because information on our other cases suggests that differences in the poverty rates between rural and urban settings are generally nominal at the midpoint of any conflict.
125. For our compiled data on poverty, please refer to appendix 2.
126. UNAIDS and UNICEF, Children on the Brink, 2002: A Joint Report on Orphan Estimates and Pro gram Strategies, July 2002, http://www.unicef.org/ publications/files/pub_children_on_the_brink_en.pdf.
127. The percentage of orphans to all children in the zero-to-14 age group was obtained for 18 cases (with the exception of Mali) in our sample. The data on orphans are presented in appendix 3.
128. The midpoint was used, to estimate orphan rates for three cases: Angola (1975-95), Angola (1996-2002), and Mozambique (1976-92). In three cases we estimated orphan rates for years before the midpoint: Senegal (1997-2001), Niger (1990-97), and the Central African Republic (2001-02). One case - Mali (1994) - was excluded because the orphans figure was reported for the year of 1995, just one year after the conflict ceased.
129. Our method does not account for the number of militarized camps or attacks in any given year or for the population of camps. Unfortunately, detailed data of this kind is currently unavailable, although one of our goals in the future project is to develop such a database.
130. Whether such a distinction is significant, however, is a subsidiary hypothesis worthy of investigation as part of our ongoing research program.
131. For reference, see Sarah Kenyon Lischer, "Refugee Involvement in Political Violence: Quantitative Evidence from 1987-1998," Working Paper, No. 26 (Cambridge, Mass.: Center for International Studies, Massachusetts Institute of Technology, June 2000), http://www.jha.ac/articlos/u026.pdf. Lischer also provided us with the dataset on which we based our calculations.
132. Appendix 4 lists all instances of camp militarization and attacks in each case. The data on refugee militarization or attacks against them for the years prior to 1988 were unavailable, reducing the data points in our two historical cases of Mozambique (1976-92) to five years from twelve and of Angola (1975-95) to eight years from thirteen. We nonetheless include these cases in our sample. The values for the access variable in our 19 cases, when compared with child soldier rates, are outlined in appendix 5.
133. With three missing values for the poverty variable, we were able to calculate results for 16 out of 19 observations. We therefore recognize that the statistical reliability of the test might be limited due to the relatively small number of observations for which data were available, and we hope to address the issue of a large N in further research.
134. In contrast to bivariate correlations, multiple regression coefficients capture the marginal effect of each independent variable on the dependent variable that is unique to the respective independent variable in the model. Hence, for example, the coefficient on the access variable expresses the rate of change in the child soldiers' variable, which could not have been associated with poverty or orphans. In this way, the multiple regression model controls for the effects of poverty and orphans.
135. If the independent variables are mutually correlated, it becomes harder to distinguish their individual effects on the dependent variable. Such cases usually result in lower precision of the coefficient estimates and hence higher standard errors.
136. R. Carter Hill, William E. Griffiths, and George G. Judge, Undergraduate Econometrics, 2d ed. (New York: Wiley, 2001), p. 190.
137. Standard errors indicate the distance from the sample regression coefficients within which the true coefficient vaiue is also likely to lie. The smaller the standard errors, the higher the accuracy of the estimated coefficient value based on the sample data. If the standard errors are small enough, relative to the coefficient value, then the coefficient is termed "significant" (i.e., estimated with a sufficient precision).
138. Among the historically most important work on the use of case studies is C. Harry Eckstein, "Case Study and Theory in Political Science," in Fred. I. Greenstein and Nelson W. Polsby, eds., Handbook of Political Science, Vol. 7: Strategies of Inquiry (Reading, Mass.: Addison-Wesley, 1975).
139. See also Mattel Dogan and Dominique Pelassy, How to Compare Nations: Strategies in Comparative Politics, 2d ed. (Chatham, N.J.: Chatham House, 1990);
140. Adam Przeworski and Henry Teune, The Logic of Comparative Social Inquiry (New York: Wiley-Interscience, 1970);
141. Alexander George, "Case Study and Theory Development: The Method of Structured, Focused Comparison," in P. Gordon Lauren, ed., Diplomacy; New Approaches in History, Theory, and Policy (New York: Free Press, 1979), pp. 43-68;
142. George and Smoke, Deterrence in American Foreign Policy, pp. 88-97;
143. and King, Keohane, and Verba, Designing Social Inquiry.
144. For armed conflicts events data, see http://www.onwar.com/aced/nation/ lay/liberia/fliberia1989.htm.
145. UNHCR, "Liberia: What Hope for Peace?"
146. For examples of human rights abuses by various rebel parties in Liberia, see Human Rights Watch, "Liberia: A Human Rights Disaster: Violations of the Laws of War by All Parties to the Conflict," October 26, 1990, http://www.hrw.org/reports/1990/liberia/.
147. "Liberia's Uneasy Peace," NewsHour, April 5, 2006, http://www.pbs.org/newshour/bb/africa/liberia/post1980_timeline.html.
148. UNHCR, "Liberia: What Hope for Peace?"
149. Norwegian Refugee Council, "Profile of Internal Displacement: Liberia," Global IDP Data-base, updated August 25, 2005, http://www.humanitarianinfo.org/liberia/mediacentre/press/doc/ Liberia%20-August%202005.pdf, p. 5.
150. Human Rights Watch, "How to Fight, How to Kill," p. 7.
151. For details of some of the worst atrocities, see Human Rights Watch, "Easy Prey," pp. 7-10.
152. "ECOMOG and Regional Peacekeeping," in Herbert Howe, Ambiguous Order: Military Forces in African States (London: Lynne Rienner, 2001), pp. 139-140.
153. Colin Scott, "Liberia: A Nation Displaced," in Roberta Cohen and Francis M. Deng, ed., The Forsaken People: Case Studies of the Internally Displaced (Washington, U.C.: Brookings, 1998), p. 114.
154. Norwegian Refugee Council, "Profile of Internal Displacement: Liberia," p. 36.
155. "ECOMOG and Regional Peacekeeping," pp. 143-144.
156. Human Rights Watch, "How to Fight, How to Kill," p. 6.
157. Ibid., "How to Fight, How to Kill, p. 15.
158. Ibid., How to Fight, How to Kill, p. 6.
159. Veronica Nmoma, "The Civil War and the Refugee Crisis in Liberia," Journal of Conflict Studies, Vol. 17, No. 1 (Spring 1997), pp. 101-125, at p. 110.
160. Scott, "Liberia," p. 113.
161. "Liberia," in Human Rights Watch World Report, 1995, http://www.hrw.org/reports/1995/WR95/AFRlCA-05.htm.
162. Nmoma, "The Civil War and the Refugee Crisis in Libya," p. 111.
163. Lischer, "Refugee Involvement in Political Violence," p. 14.
164. For further details, see "Remarks by Tarty Teh, spokesman for United Liberation Movement for Democracy in Liberia (ULIMO) before the African Subcommitlee of the House Foreign Relations Committee on the Current Situation in Liberia," November 19, 1992, http://members.aol.com/Liberia99/ UlimoSpeech.html.
165. Lischor, "Refugee Involvement in Political Violence," p. 24.
166. Ibid., Lischor, "Refugee Involvement in Political Violence," p. 24
167. Human Rights Watch, "Liberia: A Human Rights Disaster."
168. Ibid., "Liberia: A Human Rights Disaster." p. 15.
169. Scott, "Liberia," p. 114.
170. Ibid., Scott, "Liberia," p. 114.
171. Ibid., Scott, "Liberia," p. 105.
172. Norwegian Refugee Council, "Profile of Internal Displacement: Liberia," p. 10.
173. See also Scott, "Liberia," p. 108.
174. Norwegian Refugee Council, "Profile of Internal Displacement: Liberia," p. 15.
175. Ibid., "Profile of Internal Displacement: Liberia," p. 15.
176. For selection of stories, see Human Rights Watch, "Easy Prey," pp. 29, 76;
177. UNICEF, Adult Wars, Child Soldiers, pp. 23-31;
178. Yvonne H. Keairns, The Voices of Girl Child Soldiers, October 2002, http://www.geneva.quno.mfo/pdf/QUNO%20Child%20soldiers%20report.pdf.
179. See also the narratives regarding children compiled on the CSUCS website at http://www.child.-soldiers.org/childsoldiers/voices-of-young-soldiers.
180. Human Rights Watch, "Easy Prey," p. 26.
181. Human Rights Watch, "Liberia: A Human Rights Disaster."
182. Howe, Ambiguous Order.
183. Human Rights Watch, "Easy Prey," pp. 22-23.
184. According to Human Rights Watch, "During the Octopus operation in 1992, children were used by NPFL as cannon fodder. They were in the first wave of troops, and the older fighters were behind them." Human Rights Watch, "Easy Prey," p. 25.
185. Ibid., "Easy Prey," p. 24.
186. UNHCR, "Liberia: What Hope for Peace?"
187. Scott, "Liberia," p. 114.
188. Norwegian Refugee Council, "Profile of Internal Displacement: Liberia," pp. 12-13.
189. The data were obtained from different organizations and sources. See, for example, Human Rights Watch, "Easy Prey";
190. UNICEF, "The Progress of Nations, 2000," http://www.unicef.org/ pon00/;
191. CSUCS, Child Soldiers: Global Report, 2001;
192. UNICEF, The State of the World's Children, 1996;
193. and Human Rights Watch, "How to Fight, How to Kill."
194. The literature is unclear as to whether the MODEL split from LURD in early 2003 or whether it was operating as LURD's "integrated force." For instance, in "How to Fight, How to Kill," Human Rights Watch reported that MODEL split from LURD. But Jean-Hervé Jézé quel argues that MODEL joined the LURD in 2003. See Jézéquel, "Liberia: Orchestrated Chaos," In Fabrice Weissman, ed., In the Shadow of "Just Wars": Violence, Politics, and Humanitarian Action (Ithaca, N.Y.: Cornell University Press, 2004), p. 162. Other sources allowed for both possibilities.
195. See, for example, "Movement for Democracy in Liberia," GlobalSecurity.org database, http://www.globalsecurity.org/military/world/para/ model.htm.
196. Human Rights Watch, "How to Fight, How to Kill," p. 8.
197. Norwegian Refugee Council, "Profile of Internal Displacement: Liberia," p. 7.
198. "ALJA Calls for Lifting of State of Emergency in Liberia," press statement, Perspective, February 19, 2002, http://www.theperspective.org/ stateofemergency.html.
199. United Nations, "Consolidated Inter-agency Appeal, 2002: Liberia Internet," November 26, 2001, http://www.reliefweb.int/appeals/2002/files/ lib02.pdf
200. cited in ibid., "Consolidated Inter-agency Appeal, 2002: Liberia Internet," p. 106.
201. Norwegian Refugee Council, "Profile of Internal Displacement: Liberia," p. 78.
202. Ibid., "Profile of Internal Displacement: Liberia," p. 115.
203. Ibid., "Profile of Internal Displacement: Liberia" p. 7.
204. Amnesty International, "Liberian Civilians Face Human Rights Abuses At Home and Across Borders," October 2002, http://web.amnesty.org/library/ Index/engAFR340202002?OpenDocument&of=COUNTRIES\LIBERIA?OpenDocument&of= COUNTRIES\LIBERIA
205. cited in ibid., "Liberian Civilians Face Human Rights Abuses At Home and Across Borders," p. 60.
206. Norwegian Refugee Council, "Profile of Internal Displacement: Liberia," p. 37.
207. Ibid., "Profile of Internal Displacement: Liberia," p. 47.
208. One report noted, "Since the upsurge of fighting in 2000, perhaps the most pressing concern about IDPs from the north of the country has been their total lack of protection from increasingly widespread human rights abuses carried out not only by Liberian security forces but by LURD combatants as well." Ibid., "Profile of Internal Displacement: Liberia," p. 7.
209. Jézéquel, "Liberia," pp. 162-163.
210. Ibid. Jézéquel, "Liberia," pp. 162-163.
211. Ibid. Jézéquel, "Liberia," pp. 162-163.
212. The situation in Gbarpolu County, for example, had deteriorated by December 2001, as fighting forced IDPs in a camp located in Bopolu to move south and northeast. Many sought shelter at Sawmill in Bomi County. But this camp was attacked in January 2002, forcing IDPs to flee again. Likewise, in February 2002, incursions at Klay Junction forced IDPs to move toward Monrovia and Sinje in Grand Cape Mount County. Ibid., p. 25. But armed activities in Cape Mount and Bomi Counties in May 2002 resulted in a high military presence in the Sinje camps, causing panic among both refugees and IDPs living there. Ibid., p. 36. By then, the Sinje camps were home to approximately 11,000 Sierra Leonean refugees and a comparable number of IDPs. Ibid., p. 74. This pattern was repeated elsewhere: fighting in Bong County in April 2002 prompted IDPs to seek refuge near the central town of Gbarnga. Many of them sought refuge from the violence in the four camps established in the county. But the following month, the civilians were forced to flee once again when local fighting broke out and three IDP camps were forced to close. About 75,000 IDPs resided in six IDP camps in Bong County and 7,000 in a camp in Buchanan. During the fighting in June and July, many of these camps and surrounding communities were attacked by both government and rebel forces. The camps were looted, and many of the shelters were burned. Consequently, the majority of IDPs fled these camps. By the year 2002, villages and IDP camps in Lofa and Bong Counties were emptied., as their populations fled to IDP camps closer to Monrovia. Ibid., Jézéquel, "Liberia," pp. 7, 25.
213. For details regarding this effort, see Norwegian Refugee Council, "Profile of Internal Displacement: Liberia."
214. Jézéquel, "Liberia," pp. 162-163.
215. Human Rights Watch, "How to Fight, How to Kill," p. 8.
216. Jézéquel, "Liberia," p. 169.
217. Human Rights Watch, "How to Fight, How to Kill," p. 9.
218. Ibid., "How to Fight, How to Kill," p. 1.
219. Norwegian Refugee Council, "Profile of Internal Displacement: Liberia," p. 46.
220. Ibid, "Profile of Internal Displacement: Liberia," p. 6.
221. Human Rights Watch, "How to Fight, How to Kill," p. 17.
222. "How to Fight, How to Kill," Ibid., p. 17.
223. Norwegian Refugee Council, "Profile of Internal Displacement: Liberia," p. 43.
224. "Profile of Internal Displacement: Liberia," Ibid., p. 67.
225. Human Rights Watch, "How to Fight, How to Kill," p. 10
226. Ibid., "How to Fight, How to Kill," p. 11.
227. Ibid., "How to Fight, How to Kill," p. 15.
228. CSUCS, Child Soldiers Global Report, 2004, p. 77.
229. The Jartondo camp (17,000 IDPs), Wilson Center camp (25,700 IDPs), and Ricks camp (12,894 IDPs) alone sheltered. 25 percent of all Liberian IDPs and 56 percent of all IDPs settled in the region. See Norwegian Refugee Council, "Profile of Internal Displacement: Liberia."
230. Human Rights Watch, "How to Fight, How to Kill," p. 9.
231. Kenneth Bacon, director of Refugees International, interview with authors, Pittsburgh, Pennsylvania, September 23, 2004.
232. Our intended research goals include creating a database that contains accurate figures on child soldiers by year (rather than cumulative estimates by conflict); disaggregating child soldiers by different factions as well as obtaining greater details regarding individual attacks; and more data about access to refugee/IDP camps between 1975 and 1995.